| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.867992766726944, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 9.99963765490253e-06, | |
| "loss": 0.0157, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 9.998550619610118e-06, | |
| "loss": 0.0123, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 9.997463584317706e-06, | |
| "loss": 0.0152, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_loss": 0.23832739889621735, | |
| "eval_runtime": 297.425, | |
| "eval_samples_per_second": 13.271, | |
| "eval_steps_per_second": 1.661, | |
| "eval_wer": 0.17043338199023506, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 9.996376549025293e-06, | |
| "loss": 0.0167, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 9.99528951373288e-06, | |
| "loss": 0.0148, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 9.994202478440467e-06, | |
| "loss": 0.0153, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "eval_loss": 0.24725446105003357, | |
| "eval_runtime": 201.8594, | |
| "eval_samples_per_second": 19.553, | |
| "eval_steps_per_second": 2.447, | |
| "eval_wer": 0.16867166658277546, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.61, | |
| "learning_rate": 9.993115443148055e-06, | |
| "loss": 0.0163, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 9.992028407855643e-06, | |
| "loss": 0.0158, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 9.99094861946518e-06, | |
| "loss": 0.0146, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 9.989861584172766e-06, | |
| "loss": 0.0154, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "eval_loss": 0.24820923805236816, | |
| "eval_runtime": 211.4153, | |
| "eval_samples_per_second": 18.669, | |
| "eval_steps_per_second": 2.337, | |
| "eval_wer": 0.1679669804197916, | |
| "step": 1500 | |
| } | |
| ], | |
| "max_steps": 1380000, | |
| "num_train_epochs": 10000, | |
| "total_flos": 3.4217202294237536e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |